encoding.c

/**********************************************************************

  encoding.c -

  $Author$
  created at: Thu May 24 17:23:27 JST 2007

  Copyright (C) 2007 Yukihiro Matsumoto

**********************************************************************/

#include "ruby/internal/config.h"

#include <ctype.h>

#include "encindex.h"
#include "internal.h"
#include "internal/enc.h"
#include "internal/encoding.h"
#include "internal/error.h"
#include "internal/inits.h"
#include "internal/load.h"
#include "internal/object.h"
#include "internal/string.h"
#include "internal/vm.h"
#include "regenc.h"
#include "ruby/atomic.h"
#include "ruby/encoding.h"
#include "ruby/util.h"
#include "ruby/ractor.h"
#include "ruby_assert.h"
#include "vm_sync.h"
#include "ruby_atomic.h"

#ifndef ENC_DEBUG
#define ENC_DEBUG 0
#endif
#define ENC_ASSERT(expr) RUBY_ASSERT_WHEN(ENC_DEBUG, expr)
#define MUST_STRING(str) (ENC_ASSERT(RB_TYPE_P(str, T_STRING)), str)

#undef rb_ascii8bit_encindex
#undef rb_utf8_encindex
#undef rb_usascii_encindex

typedef OnigEncodingType rb_raw_encoding;

#if defined __GNUC__ && __GNUC__ >= 4
#pragma GCC visibility push(default)
int rb_enc_register(const char *name, rb_encoding *encoding);
void rb_enc_set_base(const char *name, const char *orig);
int rb_enc_set_dummy(int index);
void rb_encdb_declare(const char *name);
int rb_encdb_replicate(const char *name, const char *orig);
int rb_encdb_dummy(const char *name);
int rb_encdb_alias(const char *alias, const char *orig);
#pragma GCC visibility pop
#endif

static ID id_encoding;
VALUE rb_cEncoding;

#define ENCODING_LIST_CAPA 256
static VALUE rb_encoding_list;

struct rb_encoding_entry {
    rb_atomic_t loaded;
    const char *name;
    rb_encoding *enc;
    rb_encoding *base;
};

static struct enc_table {
    struct rb_encoding_entry list[ENCODING_LIST_CAPA];
    int count;
    st_table *names;
} global_enc_table;

static int
enc_names_free_i(st_data_t name, st_data_t idx, st_data_t args)
{
    ruby_xfree((void *)name);
    return ST_DELETE;
}

void
rb_free_global_enc_table(void)
{
    for (size_t i = 0; i < ENCODING_LIST_CAPA; i++) {
        xfree((void *)global_enc_table.list[i].enc);
    }

    st_foreach(global_enc_table.names, enc_names_free_i, (st_data_t)0);
    st_free_table(global_enc_table.names);
}

static rb_encoding *global_enc_ascii,
                   *global_enc_utf_8,
                   *global_enc_us_ascii;

#define GLOBAL_ENC_TABLE_LOCKING(tbl) \
    for (struct enc_table *tbl = &global_enc_table, **locking = &tbl; \
         locking; \
         locking = NULL) \
        RB_VM_LOCKING()


#define ENC_DUMMY_FLAG (1<<24)
#define ENC_INDEX_MASK (~(~0U<<24))

#define ENC_TO_ENCINDEX(enc) (int)((enc)->ruby_encoding_index & ENC_INDEX_MASK)
#define ENC_DUMMY_P(enc) ((enc)->ruby_encoding_index & ENC_DUMMY_FLAG)
#define ENC_SET_DUMMY(enc) ((enc)->ruby_encoding_index |= ENC_DUMMY_FLAG)

#define ENCODING_COUNT ENCINDEX_BUILTIN_MAX
#define UNSPECIFIED_ENCODING INT_MAX

#define ENCODING_NAMELEN_MAX 63
#define valid_encoding_name_p(name) ((name) && strlen(name) <= ENCODING_NAMELEN_MAX)

static const rb_data_type_t encoding_data_type = {
    "encoding",
    {0, 0, 0,},
    0, 0, RUBY_TYPED_FREE_IMMEDIATELY | RUBY_TYPED_WB_PROTECTED
};

#define is_data_encoding(obj) (RTYPEDDATA_P(obj) && RTYPEDDATA_TYPE(obj) == &encoding_data_type)
#define is_obj_encoding(obj) (RB_TYPE_P((obj), T_DATA) && is_data_encoding(obj))

int
rb_data_is_encoding(VALUE obj)
{
    return is_data_encoding(obj);
}

static VALUE
enc_new(rb_encoding *encoding)
{
    VALUE enc = TypedData_Wrap_Struct(rb_cEncoding, &encoding_data_type, (void *)encoding);
    RB_OBJ_SET_FROZEN_SHAREABLE(enc);
    return enc;
}

static void
enc_list_update(int index, rb_raw_encoding *encoding)
{
    RUBY_ASSERT(index < ENCODING_LIST_CAPA);

    VALUE list = RUBY_ATOMIC_VALUE_LOAD(rb_encoding_list);

    if (list && NIL_P(rb_ary_entry(list, index))) {
        VALUE new_list = rb_ary_dup(list);
        RBASIC_CLEAR_CLASS(new_list);
        /* initialize encoding data */
        rb_ary_store(new_list, index, enc_new(encoding));
        rb_ary_freeze(new_list);
        FL_SET_RAW(new_list, RUBY_FL_SHAREABLE);
        RUBY_ATOMIC_VALUE_SET(rb_encoding_list, new_list);
    }
}

static VALUE
enc_list_lookup(int idx)
{
    VALUE list, enc = Qnil;

    if (idx < ENCODING_LIST_CAPA) {
        list = RUBY_ATOMIC_VALUE_LOAD(rb_encoding_list);
        RUBY_ASSERT(list);
        enc = rb_ary_entry(list, idx);
    }

    if (NIL_P(enc)) {
        rb_bug("rb_enc_from_encoding_index(%d): not created yet", idx);
    }
    else {
        return enc;
    }
}

static VALUE
rb_enc_from_encoding_index(int idx)
{
    return enc_list_lookup(idx);
}

VALUE
rb_enc_from_encoding(rb_encoding *encoding)
{
    int idx;
    if (!encoding) return Qnil;
    idx = ENC_TO_ENCINDEX(encoding);
    return rb_enc_from_encoding_index(idx);
}

int
rb_enc_to_index(rb_encoding *enc)
{
    return enc ? ENC_TO_ENCINDEX(enc) : 0;
}

int
rb_enc_dummy_p(rb_encoding *enc)
{
    return ENC_DUMMY_P(enc) != 0;
}

static int
check_encoding(rb_encoding *enc)
{
    int index = rb_enc_to_index(enc);
    if (rb_enc_from_index(index) != enc)
        return -1;
    if (rb_enc_autoload_p(enc)) {
        index = rb_enc_autoload(enc);
    }
    return index;
}

static int
enc_check_encoding(VALUE obj)
{
    if (!is_obj_encoding(obj)) {
        return -1;
    }
    return check_encoding(RDATA(obj)->data);
}

NORETURN(static void not_encoding(VALUE enc));
static void
not_encoding(VALUE enc)
{
    rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE" (expected Encoding)",
             rb_obj_class(enc));
}

static rb_encoding *
must_encoding(VALUE enc)
{
    int index = enc_check_encoding(enc);
    if (index < 0) {
        not_encoding(enc);
    }
    return DATA_PTR(enc);
}

static rb_encoding *
must_encindex(int index)
{
    rb_encoding *enc = rb_enc_from_index(index);
    if (!enc) {
        rb_raise(rb_eEncodingError, "encoding index out of bound: %d",
                 index);
    }
    if (rb_enc_autoload_p(enc) && rb_enc_autoload(enc) == -1) {
        rb_loaderror("failed to load encoding (%s)",
                     rb_enc_name(enc));
    }
    if (ENC_TO_ENCINDEX(enc) != (int)(index & ENC_INDEX_MASK)) {
        rb_raise(rb_eEncodingError, "wrong encoding index %d for %s (expected %d)",
                 index, rb_enc_name(enc), ENC_TO_ENCINDEX(enc));
    }
    return enc;
}

int
rb_to_encoding_index(VALUE enc)
{
    ASSERT_vm_unlocking(); // can load encoding, so must not hold VM lock
    int idx;
    const char *name;

    idx = enc_check_encoding(enc);
    if (idx >= 0) {
        return idx;
    }
    else if (NIL_P(enc = rb_check_string_type(enc))) {
        return -1;
    }
    if (!rb_enc_asciicompat(rb_enc_get(enc))) {
        return -1;
    }
    if (!(name = rb_str_to_cstr(enc))) {
        return -1;
    }
    return rb_enc_find_index(name);
}

static const char *
name_for_encoding(volatile VALUE *enc)
{
    VALUE name = StringValue(*enc);
    const char *n;

    if (!rb_enc_asciicompat(rb_enc_get(name))) {
        rb_raise(rb_eArgError, "invalid encoding name (non ASCII)");
    }
    if (!(n = rb_str_to_cstr(name))) {
        rb_raise(rb_eArgError, "invalid encoding name (NUL byte)");
    }
    return n;
}

/* Returns encoding index or UNSPECIFIED_ENCODING */
static int
str_find_encindex(VALUE enc)
{
    int idx = rb_enc_find_index(name_for_encoding(&enc));
    RB_GC_GUARD(enc);
    return idx;
}

static int
str_to_encindex(VALUE enc)
{
    int idx = str_find_encindex(enc);
    if (idx < 0) {
        rb_raise(rb_eArgError, "unknown encoding name - %"PRIsVALUE, enc);
    }
    return idx;
}

static rb_encoding *
str_to_encoding(VALUE enc)
{
    return rb_enc_from_index(str_to_encindex(enc));
}

rb_encoding *
rb_to_encoding(VALUE enc)
{
    if (enc_check_encoding(enc) >= 0) return RDATA(enc)->data;
    return str_to_encoding(enc);
}

rb_encoding *
rb_find_encoding(VALUE enc)
{
    int idx;
    if (enc_check_encoding(enc) >= 0) return RDATA(enc)->data;
    idx = str_find_encindex(enc);
    if (idx < 0) return NULL;
    return rb_enc_from_index(idx);
}

static int
enc_table_expand(struct enc_table *enc_table, int newsize)
{
    if (newsize > ENCODING_LIST_CAPA) {
        rb_raise(rb_eEncodingError, "too many encoding (> %d)", ENCODING_LIST_CAPA);
    }
    return newsize;
}

/* Load an encoding using the values from base_encoding */
static void
enc_load_from_base(struct enc_table *enc_table, int index, rb_encoding *base_encoding)
{
    ASSERT_vm_locking();

    struct rb_encoding_entry *ent = &enc_table->list[index];

    if (ent->loaded) {
        return;
    }

    rb_raw_encoding *encoding = (rb_raw_encoding *)ent->enc;
    RUBY_ASSERT(encoding);

    // FIXME: Before the base is loaded, the encoding may be accessed
    // concurrently by other Ractors.
    // We're copying all fields from base_encoding except name and
    // ruby_encoding_index which we preserve from the original. Since these are
    // the only fields other threads should read it is likely safe despite
    // technically being a data race.
    rb_raw_encoding tmp_encoding = *base_encoding;
    tmp_encoding.name = encoding->name;
    tmp_encoding.ruby_encoding_index = encoding->ruby_encoding_index;
    *encoding = tmp_encoding;

    RUBY_ATOMIC_SET(ent->loaded, encoding->max_enc_len);
}

static int
enc_register_at(struct enc_table *enc_table, int index, const char *name, rb_encoding *base_encoding)
{
    ASSERT_vm_locking();

    struct rb_encoding_entry *ent = &enc_table->list[index];
    rb_raw_encoding *encoding;

    RUBY_ASSERT(!ent->loaded);
    RUBY_ASSERT(!ent->name);
    RUBY_ASSERT(!ent->enc);
    RUBY_ASSERT(!ent->base);

    RUBY_ASSERT(valid_encoding_name_p(name));

    ent->name = name = strdup(name);

    encoding = ZALLOC(rb_raw_encoding);
    encoding->name = name;
    encoding->ruby_encoding_index = index;
    ent->enc = encoding;

    if (st_insert(enc_table->names, (st_data_t)name, (st_data_t)index)) {
        rb_bug("encoding name was somehow registered twice");
    }

    enc_list_update(index, encoding);

    if (base_encoding) {
        enc_load_from_base(enc_table, index, base_encoding);
    }
    else {
        /* it should not be loaded yet */
        RUBY_ASSERT(!encoding->max_enc_len);
    }

    return index;
}

static int
enc_register(struct enc_table *enc_table, const char *name, rb_encoding *encoding)
{
    ASSERT_vm_locking();

    if (!valid_encoding_name_p(name)) return -1;

    int index = enc_table->count;

    enc_table->count = enc_table_expand(enc_table, index + 1);
    return enc_register_at(enc_table, index, name, encoding);
}

static void set_encoding_const(const char *, rb_encoding *);
static int enc_registered(struct enc_table *enc_table, const char *name);

static rb_encoding *
enc_from_index(struct enc_table *enc_table, int index)
{
    if (UNLIKELY(index < 0 || enc_table->count <= (index &= ENC_INDEX_MASK))) {
        return 0;
    }
    rb_encoding *enc = enc_table->list[index].enc;
    RUBY_ASSERT(ENC_TO_ENCINDEX(enc) == index);
    return enc;
}

rb_encoding *
rb_enc_from_index(int index)
{
    return enc_from_index(&global_enc_table, index);
}

int
rb_enc_register(const char *name, rb_encoding *encoding)
{
    int index;

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        index = enc_registered(enc_table, name);

        if (index >= 0) {
            rb_encoding *oldenc = enc_from_index(enc_table, index);
            if (STRCASECMP(name, rb_enc_name(oldenc))) {
                index = enc_register(enc_table, name, encoding);
            }
            else if (rb_enc_autoload_p(oldenc) || !ENC_DUMMY_P(oldenc)) {
                enc_load_from_base(enc_table, index, encoding);
            }
            else {
                rb_raise(rb_eArgError, "encoding %s is already registered", name);
            }
        }
        else {
            index = enc_register(enc_table, name, encoding);
            set_encoding_const(name, rb_enc_from_index(index));
        }
    }
    return index;
}

int
enc_registered(struct enc_table *enc_table, const char *name)
{
    ASSERT_vm_locking();
    st_data_t idx = 0;

    if (!name) return -1;
    if (!enc_table->names) return -1;
    if (st_lookup(enc_table->names, (st_data_t)name, &idx)) {
        return (int)idx;
    }
    return -1;
}

int
rb_enc_registered(const char *name)
{
    int idx;
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        idx = enc_registered(enc_table, name);
    }
    return idx;
}

void
rb_encdb_declare(const char *name)
{
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        int idx = enc_registered(enc_table, name);
        if (idx < 0) {
            idx = enc_register(enc_table, name, 0);
        }
        set_encoding_const(name, rb_enc_from_index(idx));
    }
}

static void
enc_check_addable(struct enc_table *enc_table, const char *name)
{
    if (enc_registered(enc_table, name) >= 0) {
        rb_raise(rb_eArgError, "encoding %s is already registered", name);
    }
    else if (!valid_encoding_name_p(name)) {
        rb_raise(rb_eArgError, "invalid encoding name: %s", name);
    }
}

static rb_encoding*
set_base_encoding(struct enc_table *enc_table, int index, rb_encoding *base)
{
    rb_encoding *enc = enc_table->list[index].enc;

    ASSUME(enc);
    enc_table->list[index].base = base;
    if (ENC_DUMMY_P(base)) ENC_SET_DUMMY((rb_raw_encoding *)enc);
    return enc;
}

/* for encdb.h
 * Set base encoding for encodings which are not replicas
 * but not in their own files.
 */
void
rb_enc_set_base(const char *name, const char *orig)
{
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        int idx = enc_registered(enc_table, name);
        int origidx = enc_registered(enc_table, orig);
        set_base_encoding(enc_table, idx, rb_enc_from_index(origidx));
    }
}

/* for encdb.h
 * Set encoding dummy.
 */
int
rb_enc_set_dummy(int index)
{
    rb_encoding *enc = global_enc_table.list[index].enc;
    ENC_SET_DUMMY((rb_raw_encoding *)enc);
    return index;
}

static int
enc_replicate(struct enc_table *enc_table, const char *name, rb_encoding *encoding)
{
    int idx;

    enc_check_addable(enc_table, name);
    idx = enc_register(enc_table, name, encoding);
    if (idx < 0) rb_raise(rb_eArgError, "invalid encoding name: %s", name);
    set_base_encoding(enc_table, idx, encoding);
    set_encoding_const(name, rb_enc_from_index(idx));
    return idx;
}

static int
enc_replicate_with_index(struct enc_table *enc_table, const char *name, rb_encoding *origenc, int idx)
{
    if (idx < 0) {
        idx = enc_register(enc_table, name, origenc);
    }
    else {
        enc_load_from_base(enc_table, idx, origenc);
    }
    if (idx >= 0) {
        set_base_encoding(enc_table, idx, origenc);
        set_encoding_const(name, rb_enc_from_index(idx));
    }
    else {
        rb_raise(rb_eArgError, "failed to replicate encoding");
    }
    return idx;
}

int
rb_encdb_replicate(const char *name, const char *orig)
{
    int r;

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        int origidx = enc_registered(enc_table, orig);
        int idx = enc_registered(enc_table, name);

        if (origidx < 0) {
            origidx = enc_register(enc_table, orig, 0);
        }
        r = enc_replicate_with_index(enc_table, name, rb_enc_from_index(origidx), idx);
    }

    return r;
}

int
rb_define_dummy_encoding(const char *name)
{
    int index;

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        index = enc_replicate(enc_table, name, rb_ascii8bit_encoding());
        rb_encoding *enc = enc_table->list[index].enc;
        ENC_SET_DUMMY((rb_raw_encoding *)enc);
    }

    return index;
}

int
rb_encdb_dummy(const char *name)
{
    int index;

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        index = enc_replicate_with_index(enc_table, name,
                                         rb_ascii8bit_encoding(),
                                         enc_registered(enc_table, name));
        rb_encoding *enc = enc_table->list[index].enc;
        ENC_SET_DUMMY((rb_raw_encoding *)enc);
    }

    return index;
}

/*
 * call-seq:
 *   enc.dummy? -> true or false
 *
 * Returns true for dummy encodings.
 * A dummy encoding is an encoding for which character handling is not properly
 * implemented.
 * It is used for stateful encodings.
 *
 *   Encoding::ISO_2022_JP.dummy?       #=> true
 *   Encoding::UTF_8.dummy?             #=> false
 *
 */
static VALUE
enc_dummy_p(VALUE enc)
{
    return RBOOL(ENC_DUMMY_P(must_encoding(enc)));
}

/*
 * call-seq:
 *   enc.ascii_compatible? -> true or false
 *
 * Returns whether ASCII-compatible or not.
 *
 *   Encoding::UTF_8.ascii_compatible?     #=> true
 *   Encoding::UTF_16BE.ascii_compatible?  #=> false
 *
 */
static VALUE
enc_ascii_compatible_p(VALUE enc)
{
    return RBOOL(rb_enc_asciicompat(must_encoding(enc)));
}

/*
 * Returns non-zero when the encoding is Unicode series other than UTF-7 else 0.
 */
int
rb_enc_unicode_p(rb_encoding *enc)
{
    return ONIGENC_IS_UNICODE(enc);
}

static st_data_t
enc_dup_name(st_data_t name)
{
    return (st_data_t)strdup((const char *)name);
}

/*
 * Returns copied alias name when the key is added for st_table,
 * else returns NULL.
 */
static int
enc_alias_internal(struct enc_table *enc_table, const char *alias, int idx)
{
    ASSERT_vm_locking();
    return st_insert2(enc_table->names, (st_data_t)alias, (st_data_t)idx,
                      enc_dup_name);
}

static int
enc_alias(struct enc_table *enc_table, const char *alias, int idx)
{
    if (!valid_encoding_name_p(alias)) return -1;
    if (!enc_alias_internal(enc_table, alias, idx))
        set_encoding_const(alias, enc_from_index(enc_table, idx));
    return idx;
}

int
rb_enc_alias(const char *alias, const char *orig)
{
    int idx, r;
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        enc_check_addable(enc_table, alias); // can raise
    }

    idx = rb_enc_find_index(orig);
    if (idx < 0) return -1;

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        r = enc_alias(enc_table, alias, idx);
    }

    return r;
}

int
rb_encdb_alias(const char *alias, const char *orig)
{
    int r;

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        int idx = enc_registered(enc_table, orig);

        if (idx < 0) {
            idx = enc_register(enc_table, orig, 0);
        }
        r = enc_alias(enc_table, alias, idx);
    }

    return r;
}

static void
rb_enc_init(struct enc_table *enc_table)
{
    ASSERT_vm_locking();
    enc_table_expand(enc_table, ENCODING_COUNT + 1);
    if (!enc_table->names) {
        enc_table->names = st_init_strcasetable_with_size(ENCODING_LIST_CAPA);
    }
#define OnigEncodingASCII_8BIT OnigEncodingASCII
#define ENC_REGISTER(enc) enc_register_at(enc_table, ENCINDEX_##enc, rb_enc_name(&OnigEncoding##enc), &OnigEncoding##enc)
    ENC_REGISTER(ASCII_8BIT);
    ENC_REGISTER(UTF_8);
    ENC_REGISTER(US_ASCII);
    global_enc_ascii = enc_table->list[ENCINDEX_ASCII_8BIT].enc;
    global_enc_utf_8 = enc_table->list[ENCINDEX_UTF_8].enc;
    global_enc_us_ascii = enc_table->list[ENCINDEX_US_ASCII].enc;
#undef ENC_REGISTER
#undef OnigEncodingASCII_8BIT
#define ENCDB_REGISTER(name, enc) enc_register_at(enc_table, ENCINDEX_##enc, name, NULL)
    ENCDB_REGISTER("UTF-16BE", UTF_16BE);
    ENCDB_REGISTER("UTF-16LE", UTF_16LE);
    ENCDB_REGISTER("UTF-32BE", UTF_32BE);
    ENCDB_REGISTER("UTF-32LE", UTF_32LE);
    ENCDB_REGISTER("UTF-16", UTF_16);
    ENCDB_REGISTER("UTF-32", UTF_32);
    ENCDB_REGISTER("UTF8-MAC", UTF8_MAC);

    ENCDB_REGISTER("EUC-JP", EUC_JP);
    ENCDB_REGISTER("Windows-31J", Windows_31J);
#undef ENCDB_REGISTER
    enc_table->count = ENCINDEX_BUILTIN_MAX;
}

rb_encoding *
rb_enc_get_from_index(int index)
{
    return must_encindex(index);
}

int rb_require_internal_silent(VALUE fname);

static int
load_encoding(const char *name)
{
    ASSERT_vm_unlocking();
    VALUE enclib = rb_sprintf("enc/%s.so", name);
    VALUE debug = ruby_debug;
    VALUE errinfo;
    char *s = RSTRING_PTR(enclib) + 4, *e = RSTRING_END(enclib) - 3;
    int loaded;
    int idx;

    while (s < e) {
        if (!ISALNUM(*s)) *s = '_';
        else if (ISUPPER(*s)) *s = (char)TOLOWER(*s);
        ++s;
    }
    enclib = rb_fstring(enclib);
    ruby_debug = Qfalse;
    errinfo = rb_errinfo();
    loaded = rb_require_internal_silent(enclib); // must run without VM_LOCK
    ruby_debug = debug;
    rb_set_errinfo(errinfo);

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        if (loaded < 0 || 1 < loaded) {
            idx = -1;
        }
        else if ((idx = enc_registered(enc_table, name)) < 0) {
            idx = -1;
        }
        else if (rb_enc_autoload_p(enc_table->list[idx].enc)) {
            idx = -1;
        }
    }

    return idx;
}

static int
enc_autoload_body(rb_encoding *enc)
{
    rb_encoding *base;
    int i = 0;
    ASSERT_vm_unlocking();

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        base = enc_table->list[ENC_TO_ENCINDEX(enc)].base;
    }

    if (base) {
        bool do_register = true;
        if (rb_enc_autoload_p(base)) {
            if (rb_enc_autoload(base) < 0) {
                do_register = false;
                i = -1;
            }
        }

        if (do_register) {
            GLOBAL_ENC_TABLE_LOCKING(enc_table) {
                i = ENC_TO_ENCINDEX(enc);
                enc_load_from_base(enc_table, i, base);
                RUBY_ASSERT(((rb_raw_encoding *)enc)->ruby_encoding_index == i);
            }
        }
    }
    else {
        i = -2;
    }

    return i;
}

int
rb_enc_autoload(rb_encoding *enc)
{
    ASSERT_vm_unlocking();
    int i = enc_autoload_body(enc);
    if (i == -2) {
        i = load_encoding(rb_enc_name(enc));
    }
    return i;
}

bool
rb_enc_autoload_p(rb_encoding *enc)
{
    int idx = ENC_TO_ENCINDEX(enc);
    RUBY_ASSERT(rb_enc_from_index(idx) == enc);
    return !RUBY_ATOMIC_LOAD(global_enc_table.list[idx].loaded);
}

/* Return encoding index or UNSPECIFIED_ENCODING from encoding name */
int
rb_enc_find_index(const char *name)
{
    int i;
    ASSERT_vm_unlocking(); // it needs to be unlocked so it can call `load_encoding` if necessary
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        i = enc_registered(enc_table, name);
    }
    rb_encoding *enc;

    if (i < 0) {
        i = load_encoding(name);
    }
    else if (!(enc = rb_enc_from_index(i))) {
        if (i != UNSPECIFIED_ENCODING) {
            rb_raise(rb_eArgError, "encoding %s is not registered", name);
        }
    }
    else if (rb_enc_autoload_p(enc)) {
        if (rb_enc_autoload(enc) < 0) {
            rb_warn("failed to load encoding (%s); use ASCII-8BIT instead",
                    name);
            return 0;
        }
    }
    return i;
}

int
rb_enc_find_index2(const char *name, long len)
{
    char buf[ENCODING_NAMELEN_MAX+1];

    if (len > ENCODING_NAMELEN_MAX) return -1;
    memcpy(buf, name, len);
    buf[len] = '\0';
    return rb_enc_find_index(buf);
}

rb_encoding *
rb_enc_find(const char *name)
{
    int idx = rb_enc_find_index(name);
    if (idx < 0) idx = 0;
    return rb_enc_from_index(idx);
}

static inline int
enc_capable(VALUE obj)
{
    if (SPECIAL_CONST_P(obj)) return SYMBOL_P(obj);
    switch (BUILTIN_TYPE(obj)) {
      case T_STRING:
      case T_REGEXP:
      case T_FILE:
      case T_SYMBOL:
        return TRUE;
      case T_DATA:
        if (is_data_encoding(obj)) return TRUE;
      default:
        return FALSE;
    }
}

int
rb_enc_capable(VALUE obj)
{
    return enc_capable(obj);
}

ID
rb_id_encoding(void)
{
    CONST_ID(id_encoding, "encoding");
    return id_encoding;
}

static int
enc_get_index_str(VALUE str)
{
    int i = ENCODING_GET_INLINED(str);
    if (i == ENCODING_INLINE_MAX) {
        VALUE iv;

#if 0
        iv = rb_ivar_get(str, rb_id_encoding());
        i = NUM2INT(iv);
#else
        /*
         * Tentatively, assume ASCII-8BIT, if encoding index instance
         * variable is not found.  This can happen when freeing after
         * all instance variables are removed in `obj_free`.
         */
        iv = rb_attr_get(str, rb_id_encoding());
        i = NIL_P(iv) ? ENCINDEX_ASCII_8BIT : NUM2INT(iv);
#endif
    }
    return i;
}

int
rb_enc_get_index(VALUE obj)
{
    int i = -1;
    VALUE tmp;

    if (SPECIAL_CONST_P(obj)) {
        if (!SYMBOL_P(obj)) return -1;
        obj = rb_sym2str(obj);
    }
    switch (BUILTIN_TYPE(obj)) {
      case T_STRING:
      case T_SYMBOL:
      case T_REGEXP:
        i = enc_get_index_str(obj);
        break;
      case T_FILE:
        tmp = rb_funcallv(obj, rb_intern("internal_encoding"), 0, 0);
        if (NIL_P(tmp)) {
            tmp = rb_funcallv(obj, rb_intern("external_encoding"), 0, 0);
        }
        if (is_obj_encoding(tmp)) {
            i = enc_check_encoding(tmp);
        }
        break;
      case T_DATA:
        if (is_data_encoding(obj)) {
            i = enc_check_encoding(obj);
        }
        break;
      default:
        break;
    }
    return i;
}

static void
enc_set_index(VALUE obj, int idx)
{
    if (!enc_capable(obj)) {
        rb_raise(rb_eArgError, "cannot set encoding on non-encoding capable object");
    }

    if (idx < ENCODING_INLINE_MAX) {
        ENCODING_SET_INLINED(obj, idx);
        return;
    }
    ENCODING_SET_INLINED(obj, ENCODING_INLINE_MAX);
    rb_ivar_set(obj, rb_id_encoding(), INT2NUM(idx));
}

void
rb_enc_raw_set(VALUE obj, rb_encoding *enc)
{
    RUBY_ASSERT(enc_capable(obj));

    int idx = enc ? ENC_TO_ENCINDEX(enc) : 0;

    if (idx < ENCODING_INLINE_MAX) {
        ENCODING_SET_INLINED(obj, idx);
        return;
    }
    ENCODING_SET_INLINED(obj, ENCODING_INLINE_MAX);
    rb_ivar_set(obj, rb_id_encoding(), INT2NUM(idx));
}

void
rb_enc_set_index(VALUE obj, int idx)
{
    rb_check_frozen(obj);
    must_encindex(idx);
    enc_set_index(obj, idx);
}

VALUE
rb_enc_associate_index(VALUE obj, int idx)
{
    rb_encoding *enc;
    int oldidx, oldtermlen, termlen;

    rb_check_frozen(obj);
    oldidx = rb_enc_get_index(obj);
    if (oldidx == idx)
        return obj;
    if (SPECIAL_CONST_P(obj)) {
        rb_raise(rb_eArgError, "cannot set encoding");
    }
    enc = must_encindex(idx);
    if (!ENC_CODERANGE_ASCIIONLY(obj) ||
        !rb_enc_asciicompat(enc)) {
        ENC_CODERANGE_CLEAR(obj);
    }
    termlen = rb_enc_mbminlen(enc);
    oldtermlen = rb_enc_mbminlen(rb_enc_from_index(oldidx));
    if (oldtermlen != termlen && RB_TYPE_P(obj, T_STRING)) {
        rb_str_change_terminator_length(obj, oldtermlen, termlen);
    }
    enc_set_index(obj, idx);
    return obj;
}

VALUE
rb_enc_associate(VALUE obj, rb_encoding *enc)
{
    return rb_enc_associate_index(obj, rb_enc_to_index(enc));
}

rb_encoding*
rb_enc_get(VALUE obj)
{
    return rb_enc_from_index(rb_enc_get_index(obj));
}

const char *
rb_enc_inspect_name(rb_encoding *enc)
{
    if (enc == global_enc_ascii) {
        return "BINARY (ASCII-8BIT)";
    }
    return enc->name;
}

static rb_encoding*
rb_encoding_check(rb_encoding* enc, VALUE str1, VALUE str2)
{
    if (!enc)
        rb_raise(rb_eEncCompatError, "incompatible character encodings: %s and %s",
                 rb_enc_inspect_name(rb_enc_get(str1)),
                 rb_enc_inspect_name(rb_enc_get(str2)));
    return enc;
}

static rb_encoding* enc_compatible_str(VALUE str1, VALUE str2);

rb_encoding*
rb_enc_check_str(VALUE str1, VALUE str2)
{
    rb_encoding *enc = enc_compatible_str(MUST_STRING(str1), MUST_STRING(str2));
    return rb_encoding_check(enc, str1, str2);
}

rb_encoding*
rb_enc_check(VALUE str1, VALUE str2)
{
    rb_encoding *enc = rb_enc_compatible(str1, str2);
    return rb_encoding_check(enc, str1, str2);
}

static rb_encoding*
enc_compatible_latter(VALUE str1, VALUE str2, int idx1, int idx2)
{
    if (idx1 < 0 || idx2 < 0)
        return 0;

    if (idx1 == idx2) {
        return rb_enc_from_index(idx1);
    }

    int isstr1, isstr2;
    rb_encoding *enc1 = rb_enc_from_index(idx1);
    rb_encoding *enc2 = rb_enc_from_index(idx2);

    isstr2 = RB_TYPE_P(str2, T_STRING);
    if (isstr2 && RSTRING_LEN(str2) == 0)
        return enc1;
    isstr1 = RB_TYPE_P(str1, T_STRING);
    if (isstr1 && isstr2 && RSTRING_LEN(str1) == 0)
        return (rb_enc_asciicompat(enc1) && rb_enc_str_asciionly_p(str2)) ? enc1 : enc2;
    if (!rb_enc_asciicompat(enc1) || !rb_enc_asciicompat(enc2)) {
        return 0;
    }

    /* objects whose encoding is the same of contents */
    if (!isstr2 && idx2 == ENCINDEX_US_ASCII)
        return enc1;
    if (!isstr1 && idx1 == ENCINDEX_US_ASCII)
        return enc2;

    if (!isstr1) {
        VALUE tmp = str1;
        int idx0 = idx1;
        str1 = str2;
        str2 = tmp;
        idx1 = idx2;
        idx2 = idx0;
        idx0 = isstr1;
        isstr1 = isstr2;
        isstr2 = idx0;
    }
    if (isstr1) {
        int cr1, cr2;

        cr1 = rb_enc_str_coderange(str1);
        if (isstr2) {
            cr2 = rb_enc_str_coderange(str2);
            if (cr1 != cr2) {
                /* may need to handle ENC_CODERANGE_BROKEN */
                if (cr1 == ENC_CODERANGE_7BIT) return enc2;
                if (cr2 == ENC_CODERANGE_7BIT) return enc1;
            }
            if (cr2 == ENC_CODERANGE_7BIT) {
                return enc1;
            }
        }
        if (cr1 == ENC_CODERANGE_7BIT)
            return enc2;
    }
    return 0;
}

static rb_encoding*
enc_compatible_str(VALUE str1, VALUE str2)
{
    int idx1 = enc_get_index_str(str1);
    int idx2 = enc_get_index_str(str2);

    return enc_compatible_latter(str1, str2, idx1, idx2);
}

rb_encoding*
rb_enc_compatible(VALUE str1, VALUE str2)
{
    int idx1 = rb_enc_get_index(str1);
    int idx2 = rb_enc_get_index(str2);

    return enc_compatible_latter(str1, str2, idx1, idx2);
}

void
rb_enc_copy(VALUE obj1, VALUE obj2)
{
    rb_enc_associate_index(obj1, rb_enc_get_index(obj2));
}


/*
 *  call-seq:
 *     encoding -> encoding
 *
 *  Returns an Encoding object that represents the encoding of +self+;
 *  see {Encodings}[rdoc-ref:encodings.rdoc].
 *
 *  Related: see {Querying}[rdoc-ref:String@Querying].
 */

VALUE
rb_obj_encoding(VALUE obj)
{
    int idx = rb_enc_get_index(obj);
    if (idx < 0) {
        rb_raise(rb_eTypeError, "unknown encoding");
    }
    return rb_enc_from_encoding_index(idx & ENC_INDEX_MASK);
}

int
rb_enc_fast_mbclen(const char *p, const char *e, rb_encoding *enc)
{
    return ONIGENC_MBC_ENC_LEN(enc, (UChar*)p, (UChar*)e);
}

int
rb_enc_mbclen(const char *p, const char *e, rb_encoding *enc)
{
    int n = ONIGENC_PRECISE_MBC_ENC_LEN(enc, (UChar*)p, (UChar*)e);
    if (MBCLEN_CHARFOUND_P(n) && MBCLEN_CHARFOUND_LEN(n) <= e-p)
        return MBCLEN_CHARFOUND_LEN(n);
    else {
        int min = rb_enc_mbminlen(enc);
        return min <= e-p ? min : (int)(e-p);
    }
}

int
rb_enc_precise_mbclen(const char *p, const char *e, rb_encoding *enc)
{
    int n;
    if (e <= p)
        return ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE(1);
    n = ONIGENC_PRECISE_MBC_ENC_LEN(enc, (UChar*)p, (UChar*)e);
    if (e-p < n)
        return ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE(n-(int)(e-p));
    return n;
}

int
rb_enc_ascget(const char *p, const char *e, int *len, rb_encoding *enc)
{
    unsigned int c;
    int l;
    if (e <= p)
        return -1;
    if (rb_enc_asciicompat(enc)) {
        c = (unsigned char)*p;
        if (!ISASCII(c))
            return -1;
        if (len) *len = 1;
        return c;
    }
    l = rb_enc_precise_mbclen(p, e, enc);
    if (!MBCLEN_CHARFOUND_P(l))
        return -1;
    c = rb_enc_mbc_to_codepoint(p, e, enc);
    if (!rb_enc_isascii(c, enc))
        return -1;
    if (len) *len = l;
    return c;
}

unsigned int
rb_enc_codepoint_len(const char *p, const char *e, int *len_p, rb_encoding *enc)
{
    int r;
    if (e <= p)
        rb_raise(rb_eArgError, "empty string");
    r = rb_enc_precise_mbclen(p, e, enc);
    if (!MBCLEN_CHARFOUND_P(r)) {
        rb_raise(rb_eArgError, "invalid byte sequence in %s", rb_enc_name(enc));
    }
    if (len_p) *len_p = MBCLEN_CHARFOUND_LEN(r);
    return rb_enc_mbc_to_codepoint(p, e, enc);
}

int
rb_enc_codelen(int c, rb_encoding *enc)
{
    int n = ONIGENC_CODE_TO_MBCLEN(enc,c);
    if (n == 0) {
        rb_raise(rb_eArgError, "invalid codepoint 0x%x in %s", c, rb_enc_name(enc));
    }
    return n;
}

int
rb_enc_toupper(int c, rb_encoding *enc)
{
    return (ONIGENC_IS_ASCII_CODE(c)?ONIGENC_ASCII_CODE_TO_UPPER_CASE(c):(c));
}

int
rb_enc_tolower(int c, rb_encoding *enc)
{
    return (ONIGENC_IS_ASCII_CODE(c)?ONIGENC_ASCII_CODE_TO_LOWER_CASE(c):(c));
}

/*
 * call-seq:
 *   enc.inspect -> string
 *
 * Returns a string which represents the encoding for programmers.
 *
 *   Encoding::UTF_8.inspect       #=> "#<Encoding:UTF-8>"
 *   Encoding::ISO_2022_JP.inspect #=> "#<Encoding:ISO-2022-JP (dummy)>"
 */
static VALUE
enc_inspect(VALUE self)
{
    rb_encoding *enc;

    if (!is_data_encoding(self)) {
        not_encoding(self);
    }
    if (!(enc = DATA_PTR(self)) || rb_enc_from_index(rb_enc_to_index(enc)) != enc) {
        rb_raise(rb_eTypeError, "broken Encoding");
    }

    return rb_enc_sprintf(rb_usascii_encoding(),
                          "#<%"PRIsVALUE":%s%s%s>", rb_obj_class(self),
                          rb_enc_inspect_name(enc),
                          (ENC_DUMMY_P(enc) ? " (dummy)" : ""),
                          rb_enc_autoload_p(enc) ? " (autoload)" : "");
}

/*
 * call-seq:
 *   enc.name -> string
 *   enc.to_s -> string
 *
 * Returns the name of the encoding.
 *
 *   Encoding::UTF_8.name      #=> "UTF-8"
 */
static VALUE
enc_name(VALUE self)
{
    return rb_fstring_cstr(rb_enc_name((rb_encoding*)DATA_PTR(self)));
}

static int
enc_names_i(st_data_t name, st_data_t idx, st_data_t args)
{
    VALUE *arg = (VALUE *)args;

    if ((int)idx == (int)arg[0]) {
        VALUE str = rb_interned_str_cstr((char *)name);
        rb_ary_push(arg[1], str);
    }
    return ST_CONTINUE;
}

/*
 * call-seq:
 *   enc.names -> array
 *
 * Returns the list of name and aliases of the encoding.
 *
 *   Encoding::WINDOWS_31J.names  #=> ["Windows-31J", "CP932", "csWindows31J", "SJIS", "PCK"]
 */
static VALUE
enc_names(VALUE self)
{
    VALUE args[2];

    args[0] = (VALUE)rb_to_encoding_index(self);
    args[1] = rb_ary_new2(0);

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        st_foreach(enc_table->names, enc_names_i, (st_data_t)args);
    }
    return args[1];
}

/*
 * call-seq:
 *   Encoding.list -> [enc1, enc2, ...]
 *
 * Returns the list of loaded encodings.
 *
 *   Encoding.list
 *   #=> [#<Encoding:ASCII-8BIT>, #<Encoding:UTF-8>,
 *         #<Encoding:ISO-2022-JP (dummy)>]
 *
 *   Encoding.find("US-ASCII")
 *   #=> #<Encoding:US-ASCII>
 *
 *   Encoding.list
 *   #=> [#<Encoding:ASCII-8BIT>, #<Encoding:UTF-8>,
 *         #<Encoding:US-ASCII>, #<Encoding:ISO-2022-JP (dummy)>]
 *
 */
static VALUE
enc_list(VALUE klass)
{
    VALUE list = RUBY_ATOMIC_VALUE_LOAD(rb_encoding_list);
    return rb_ary_dup(list);
}

/*
 * call-seq:
 *   Encoding.find(string) -> enc
 *
 * Search the encoding with specified <i>name</i>.
 * <i>name</i> should be a string.
 *
 *   Encoding.find("US-ASCII")  #=> #<Encoding:US-ASCII>
 *
 * Names which this method accept are encoding names and aliases
 * including following special aliases
 *
 * "external"::   default external encoding
 * "internal"::   default internal encoding
 * "locale"::     locale encoding
 * "filesystem":: filesystem encoding
 *
 * An ArgumentError is raised when no encoding with <i>name</i>.
 * Only <code>Encoding.find("internal")</code> however returns nil
 * when no encoding named "internal", in other words, when Ruby has no
 * default internal encoding.
 */
static VALUE
enc_find(VALUE klass, VALUE enc)
{
    int idx;
    if (is_obj_encoding(enc))
        return enc;
    idx = str_to_encindex(enc);
    if (idx == UNSPECIFIED_ENCODING) return Qnil;
    return rb_enc_from_encoding_index(idx);
}

/*
 * call-seq:
 *   Encoding.compatible?(obj1, obj2) -> enc or nil
 *
 * Checks the compatibility of two objects.
 *
 * If the objects are both strings they are compatible when they are
 * concatenatable.  The encoding of the concatenated string will be returned
 * if they are compatible, nil if they are not.
 *
 *   Encoding.compatible?("\xa1".force_encoding("iso-8859-1"), "b")
 *   #=> #<Encoding:ISO-8859-1>
 *
 *   Encoding.compatible?(
 *     "\xa1".force_encoding("iso-8859-1"),
 *     "\xa1\xa1".force_encoding("euc-jp"))
 *   #=> nil
 *
 * If the objects are non-strings their encodings are compatible when they
 * have an encoding and:
 * * Either encoding is US-ASCII compatible
 * * One of the encodings is a 7-bit encoding
 *
 */
static VALUE
enc_compatible_p(VALUE klass, VALUE str1, VALUE str2)
{
    rb_encoding *enc;

    if (!enc_capable(str1)) return Qnil;
    if (!enc_capable(str2)) return Qnil;
    enc = rb_enc_compatible(str1, str2);
    if (!enc) return Qnil;
    return rb_enc_from_encoding(enc);
}

NORETURN(static VALUE enc_s_alloc(VALUE klass));
/* :nodoc: */
static VALUE
enc_s_alloc(VALUE klass)
{
    rb_undefined_alloc(klass);
    UNREACHABLE_RETURN(Qnil);
}

/* :nodoc: */
static VALUE
enc_dump(int argc, VALUE *argv, VALUE self)
{
    rb_check_arity(argc, 0, 1);
    return enc_name(self);
}

/* :nodoc: */
static VALUE
enc_load(VALUE klass, VALUE str)
{
    return str;
}

/* :nodoc: */
static VALUE
enc_m_loader(VALUE klass, VALUE str)
{
    return enc_find(klass, str);
}

rb_encoding *
rb_ascii8bit_encoding(void)
{
    return global_enc_ascii;
}

int
rb_ascii8bit_encindex(void)
{
    return ENCINDEX_ASCII_8BIT;
}

rb_encoding *
rb_utf8_encoding(void)
{
    return global_enc_utf_8;
}

int
rb_utf8_encindex(void)
{
    return ENCINDEX_UTF_8;
}

rb_encoding *
rb_usascii_encoding(void)
{
    return global_enc_us_ascii;
}

int
rb_usascii_encindex(void)
{
    return ENCINDEX_US_ASCII;
}

int rb_locale_charmap_index(void);

int
rb_locale_encindex(void)
{
    // `rb_locale_charmap_index` can call `enc_find_index`, which can
    // load an encoding. This needs to be done without VM lock held.
    ASSERT_vm_unlocking();
    int idx = rb_locale_charmap_index();

    if (idx < 0) idx = ENCINDEX_UTF_8;

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        if (enc_registered(enc_table, "locale") < 0) {
# if defined _WIN32
            void Init_w32_codepage(void);
            Init_w32_codepage();
# endif
            GLOBAL_ENC_TABLE_LOCKING(enc_table) {
                enc_alias_internal(enc_table, "locale", idx);
            }
        }
    }

    return idx;
}

rb_encoding *
rb_locale_encoding(void)
{
    return rb_enc_from_index(rb_locale_encindex());
}

int
rb_filesystem_encindex(void)
{
    int idx;
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        idx = enc_registered(enc_table, "filesystem");
    }
    if (idx < 0) idx = ENCINDEX_ASCII_8BIT;
    return idx;
}

rb_encoding *
rb_filesystem_encoding(void)
{
    return rb_enc_from_index(rb_filesystem_encindex());
}

struct default_encoding {
    int index;			/* -2 => not yet set, -1 => nil */
    rb_encoding *enc;
};

static struct default_encoding default_external = {0};

static int
enc_set_default_encoding(struct default_encoding *def, VALUE encoding, const char *name)
{
    int overridden = FALSE;

    if (def->index != -2)
        /* Already set */
        overridden = TRUE;

    int index = 0;
    if (!NIL_P(encoding)) {
        enc_check_encoding(encoding); // loads it if necessary. Needs to be done outside of VM lock.
        index = rb_enc_to_index(rb_to_encoding(encoding));
    }

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        if (NIL_P(encoding)) {
            def->index = -1;
            def->enc = 0;
            char *name_dup = strdup(name);

            st_data_t existing_name = (st_data_t)name_dup;
            if (st_delete(enc_table->names, &existing_name, NULL)) {
                xfree((void *)existing_name);
            }

            st_insert(enc_table->names, (st_data_t)name_dup,
                      (st_data_t)UNSPECIFIED_ENCODING);
        }
        else {
            def->index = index;
            def->enc = 0;
            enc_alias_internal(enc_table, name, def->index);
        }

        if (def == &default_external) {
            enc_alias_internal(enc_table, "filesystem", Init_enc_set_filesystem_encoding());
        }
    }

    return overridden;
}

rb_encoding *
rb_default_external_encoding(void)
{
    if (default_external.enc) return default_external.enc;

    if (default_external.index >= 0) {
        default_external.enc = rb_enc_from_index(default_external.index);
        return default_external.enc;
    }
    else {
        return rb_locale_encoding();
    }
}

VALUE
rb_enc_default_external(void)
{
    return rb_enc_from_encoding(rb_default_external_encoding());
}

/*
 * call-seq:
 *   Encoding.default_external -> enc
 *
 * Returns default external encoding.
 *
 * The default external encoding is used by default for strings created from
 * the following locations:
 *
 * * CSV
 * * File data read from disk
 * * SDBM
 * * StringIO
 * * Zlib::GzipReader
 * * Zlib::GzipWriter
 * * String#inspect
 * * Regexp#inspect
 *
 * While strings created from these locations will have this encoding, the
 * encoding may not be valid.  Be sure to check String#valid_encoding?.
 *
 * File data written to disk will be transcoded to the default external
 * encoding when written, if default_internal is not nil.
 *
 * The default external encoding is initialized by the -E option.
 * If -E isn't set, it is initialized to UTF-8 on Windows and the locale on
 * other operating systems.
 */
static VALUE
get_default_external(VALUE klass)
{
    return rb_enc_default_external();
}

void
rb_enc_set_default_external(VALUE encoding)
{
    if (NIL_P(encoding)) {
        rb_raise(rb_eArgError, "default external can not be nil");
    }
    enc_set_default_encoding(&default_external, encoding,
                            "external");
}

/*
 * call-seq:
 *   Encoding.default_external = enc
 *
 * Sets default external encoding.  You should not set
 * Encoding::default_external in ruby code as strings created before changing
 * the value may have a different encoding from strings created after the value
 * was changed., instead you should use <tt>ruby -E</tt> to invoke ruby with
 * the correct default_external.
 *
 * See Encoding::default_external for information on how the default external
 * encoding is used.
 */
static VALUE
set_default_external(VALUE klass, VALUE encoding)
{
    rb_warning("setting Encoding.default_external");
    rb_enc_set_default_external(encoding);
    return encoding;
}

static struct default_encoding default_internal = {-2};

rb_encoding *
rb_default_internal_encoding(void)
{
    if (!default_internal.enc && default_internal.index >= 0) {
        default_internal.enc = rb_enc_from_index(default_internal.index);
    }
    return default_internal.enc; /* can be NULL */
}

VALUE
rb_enc_default_internal(void)
{
    /* Note: These functions cope with default_internal not being set */
    return rb_enc_from_encoding(rb_default_internal_encoding());
}

/*
 * call-seq:
 *   Encoding.default_internal -> enc
 *
 * Returns default internal encoding.  Strings will be transcoded to the
 * default internal encoding in the following places if the default internal
 * encoding is not nil:
 *
 * * CSV
 * * Etc.sysconfdir and Etc.systmpdir
 * * File data read from disk
 * * File names from Dir
 * * Integer#chr
 * * String#inspect and Regexp#inspect
 * * Strings returned from Readline
 * * Strings returned from SDBM
 * * Time#zone
 * * Values from ENV
 * * Values in ARGV including $PROGRAM_NAME
 *
 * Additionally String#encode and String#encode! use the default internal
 * encoding if no encoding is given.
 *
 * The script encoding (__ENCODING__), not default_internal, is used as the
 * encoding of created strings.
 *
 * Encoding::default_internal is initialized with -E option or nil otherwise.
 */
static VALUE
get_default_internal(VALUE klass)
{
    return rb_enc_default_internal();
}

void
rb_enc_set_default_internal(VALUE encoding)
{
    enc_set_default_encoding(&default_internal, encoding,
                            "internal");
}

/*
 * call-seq:
 *   Encoding.default_internal = enc or nil
 *
 * Sets default internal encoding or removes default internal encoding when
 * passed nil.  You should not set Encoding::default_internal in ruby code as
 * strings created before changing the value may have a different encoding
 * from strings created after the change.  Instead you should use
 * <tt>ruby -E</tt> to invoke ruby with the correct default_internal.
 *
 * See Encoding::default_internal for information on how the default internal
 * encoding is used.
 */
static VALUE
set_default_internal(VALUE klass, VALUE encoding)
{
    rb_warning("setting Encoding.default_internal");
    rb_enc_set_default_internal(encoding);
    return encoding;
}

static void
set_encoding_const(const char *name, rb_encoding *enc)
{
    VALUE encoding = rb_enc_from_encoding(enc);
    char *s = (char *)name;
    int haslower = 0, hasupper = 0, valid = 0;

    if (ISDIGIT(*s)) return;
    if (ISUPPER(*s)) {
        hasupper = 1;
        while (*++s && (ISALNUM(*s) || *s == '_')) {
            if (ISLOWER(*s)) haslower = 1;
        }
    }
    if (!*s) {
        if (s - name > ENCODING_NAMELEN_MAX) return;
        valid = 1;
        rb_define_const(rb_cEncoding, name, encoding);
    }
    if (!valid || haslower) {
        size_t len = s - name;
        if (len > ENCODING_NAMELEN_MAX) return;
        if (!haslower || !hasupper) {
            do {
                if (ISLOWER(*s)) haslower = 1;
                if (ISUPPER(*s)) hasupper = 1;
            } while (*++s && (!haslower || !hasupper));
            len = s - name;
        }
        len += strlen(s);
        if (len++ > ENCODING_NAMELEN_MAX) return;
        MEMCPY(s = ALLOCA_N(char, len), name, char, len);
        name = s;
        if (!valid) {
            if (ISLOWER(*s)) *s = ONIGENC_ASCII_CODE_TO_UPPER_CASE((int)*s);
            for (; *s; ++s) {
                if (!ISALNUM(*s)) *s = '_';
            }
            if (hasupper) {
                rb_define_const(rb_cEncoding, name, encoding);
            }
        }
        if (haslower) {
            for (s = (char *)name; *s; ++s) {
                if (ISLOWER(*s)) *s = ONIGENC_ASCII_CODE_TO_UPPER_CASE((int)*s);
            }
            rb_define_const(rb_cEncoding, name, encoding);
        }
    }
}

static int
rb_enc_name_list_i(st_data_t name, st_data_t idx, st_data_t arg)
{
    VALUE ary = (VALUE)arg;
    VALUE str = rb_interned_str_cstr((char *)name);
    rb_ary_push(ary, str);
    return ST_CONTINUE;
}

/*
 * call-seq:
 *   Encoding.name_list -> ["enc1", "enc2", ...]
 *
 * Returns the list of available encoding names.
 *
 *   Encoding.name_list
 *   #=> ["US-ASCII", "ASCII-8BIT", "UTF-8",
 *         "ISO-8859-1", "Shift_JIS", "EUC-JP",
 *         "Windows-31J",
 *         "BINARY", "CP932", "eucJP"]
 *
 */

static VALUE
rb_enc_name_list(VALUE klass)
{
    VALUE ary;
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        ary = rb_ary_new2(enc_table->names->num_entries);
        st_foreach(enc_table->names, rb_enc_name_list_i, (st_data_t)ary);
    }
    return ary;
}

static int
rb_enc_aliases_enc_i(st_data_t name, st_data_t orig, st_data_t arg)
{
    VALUE *p = (VALUE *)arg;
    VALUE aliases = p[0], ary = p[1];
    int idx = (int)orig;
    VALUE key, str = rb_ary_entry(ary, idx);

    if (NIL_P(str)) {
        rb_encoding *enc = rb_enc_from_index(idx);

        if (!enc) return ST_CONTINUE;
        if (STRCASECMP((char*)name, rb_enc_name(enc)) == 0) {
            return ST_CONTINUE;
        }
        str = rb_fstring_cstr(rb_enc_name(enc));
        rb_ary_store(ary, idx, str);
    }
    key = rb_interned_str_cstr((char *)name);
    rb_hash_aset(aliases, key, str);
    return ST_CONTINUE;
}

/*
 * call-seq:
 *   Encoding.aliases -> {"alias1" => "orig1", "alias2" => "orig2", ...}
 *
 * Returns the hash of available encoding alias and original encoding name.
 *
 *   Encoding.aliases
 *   #=> {"BINARY"=>"ASCII-8BIT", "ASCII"=>"US-ASCII", "ANSI_X3.4-1968"=>"US-ASCII",
 *         "SJIS"=>"Windows-31J", "eucJP"=>"EUC-JP", "CP932"=>"Windows-31J"}
 *
 */

static VALUE
rb_enc_aliases(VALUE klass)
{
    VALUE aliases[2];
    aliases[0] = rb_hash_new();
    aliases[1] = rb_ary_new();

    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        st_foreach(enc_table->names, rb_enc_aliases_enc_i, (st_data_t)aliases);
    }

    return aliases[0];
}

/*
 * An \Encoding instance represents a character encoding usable in Ruby.
 * It is defined as a constant under the \Encoding namespace.
 * It has a name and, optionally, aliases:
 *
 *   Encoding::US_ASCII.name  # => "US-ASCII"
 *   Encoding::US_ASCII.names # => ["US-ASCII", "ASCII", "ANSI_X3.4-1968", "646"]
 *
 * A Ruby method that accepts an encoding as an argument will accept:
 *
 * - An \Encoding object.
 * - The name of an encoding.
 * - An alias for an encoding name.
 *
 * These are equivalent:
 *
 *   'foo'.encode(Encoding::US_ASCII) # Encoding object.
 *   'foo'.encode('US-ASCII')         # Encoding name.
 *   'foo'.encode('ASCII')            # Encoding alias.
 *
 * For a full discussion of encodings and their uses,
 * see {the Encodings document}[rdoc-ref:encodings.rdoc].
 *
 * Encoding::ASCII_8BIT is a special-purpose encoding that is usually used for
 * a string of bytes, not a string of characters.
 * But as the name indicates, its characters in the ASCII range
 * are considered as ASCII characters.
 * This is useful when you use other ASCII-compatible encodings.
 *
 */

void
Init_Encoding(void)
{
    VALUE list;
    int i;

    rb_cEncoding = rb_define_class("Encoding", rb_cObject);
    rb_define_alloc_func(rb_cEncoding, enc_s_alloc);
    rb_undef_method(CLASS_OF(rb_cEncoding), "new");
    rb_define_method(rb_cEncoding, "to_s", enc_name, 0);
    rb_define_method(rb_cEncoding, "inspect", enc_inspect, 0);
    rb_define_method(rb_cEncoding, "name", enc_name, 0);
    rb_define_method(rb_cEncoding, "names", enc_names, 0);
    rb_define_method(rb_cEncoding, "dummy?", enc_dummy_p, 0);
    rb_define_method(rb_cEncoding, "ascii_compatible?", enc_ascii_compatible_p, 0);
    rb_define_singleton_method(rb_cEncoding, "list", enc_list, 0);
    rb_define_singleton_method(rb_cEncoding, "name_list", rb_enc_name_list, 0);
    rb_define_singleton_method(rb_cEncoding, "aliases", rb_enc_aliases, 0);
    rb_define_singleton_method(rb_cEncoding, "find", enc_find, 1);
    rb_define_singleton_method(rb_cEncoding, "compatible?", enc_compatible_p, 2);

    rb_define_method(rb_cEncoding, "_dump", enc_dump, -1);
    rb_define_singleton_method(rb_cEncoding, "_load", enc_load, 1);

    rb_define_singleton_method(rb_cEncoding, "default_external", get_default_external, 0);
    rb_define_singleton_method(rb_cEncoding, "default_external=", set_default_external, 1);
    rb_define_singleton_method(rb_cEncoding, "default_internal", get_default_internal, 0);
    rb_define_singleton_method(rb_cEncoding, "default_internal=", set_default_internal, 1);
    rb_define_singleton_method(rb_cEncoding, "locale_charmap", rb_locale_charmap, 0); /* in localeinit.c */

    struct enc_table *enc_table = &global_enc_table;

    rb_gc_register_address(&rb_encoding_list);
    list = rb_encoding_list = rb_ary_new2(ENCODING_LIST_CAPA);
    RBASIC_CLEAR_CLASS(list);

    for (i = 0; i < enc_table->count; ++i) {
        rb_ary_push(list, enc_new(enc_table->list[i].enc));
    }

    rb_marshal_define_compat(rb_cEncoding, Qnil, 0, enc_m_loader);
}

void
Init_unicode_version(void)
{
    extern const char onigenc_unicode_version_string[];

    VALUE str = rb_usascii_str_new_static(onigenc_unicode_version_string,
                                          strlen(onigenc_unicode_version_string));
    OBJ_FREEZE(str);
    /* The supported Unicode version. */
    rb_define_const(rb_cEncoding, "UNICODE_VERSION", str);
}

void
Init_encodings(void)
{
    rb_enc_init(&global_enc_table);
}

/* locale insensitive ctype functions */

void
rb_enc_foreach_name(int (*func)(st_data_t name, st_data_t idx, st_data_t arg), st_data_t arg)
{
    GLOBAL_ENC_TABLE_LOCKING(enc_table) {
        st_foreach(enc_table->names, func, arg);
    }
}